Magnetrons are well known in the art and have long served as highly efficient sources of microwave energy. For example, magnetrons are commonly employed in microwave ovens to generate sufficient microwave energy for heating and cooking various foods. The use of magnetrons is desirable in that they operate with high efficiency, thus avoiding high costs associated with excess power consumption, heat dissipation, etc.
Conventional microwave magnetrons employ a constant electric and magnetic field to produce a rotating electron space charge. The electron space charge interacts with a plurality of microwave resonant cavities to generate microwave radiation. Conventional magnetrons are efficient generators of microwave energy for frequencies in the 1 to 10 GHz region. At higher frequencies, the maximum output power drops and the required electric and magnetic field strength increases (at higher frequencies the resonant cavities become proportionally smaller). The practical upper frequency limit for conventional magnetron designs is about 100 GHz at about 1 Watt (W) of continuous power. By comparison, at 1 GHz, conventional magnetrons can produce several kilowatts of continuous power. In short pulses, most magnetron designs can produce peak powers 1000 times higher than their maximum continuous power levels. In pulse operation, multi-megawatt power levels are possible in the 1 to 10 GHz range.
Conventional magnetrons employ anodes which have a plurality of resonant cavities arranged around a cylindrical cathode. The resonant cavities typically number from six to twenty. They may be shaped as hole and slot-keyhole structures or as straight-sided pie-shaped structures. FIGS. 1A-1C illustrate several conventional magnetron anode designs, namely, the slot-keyhole, the straight-sided pie-shaped structure and the rising sun structure (i.e., an anode with resonant cavities having varying dimensions), respectively.
Mode control is an important issue in magnetron operation. A mode is a collective oscillation of all of the resonant cavities. In a single mode, all of the cavities may oscillate at substantially the same frequency but with some phase difference between adjacent cavities. The most desirable mode of operation occurs when adjacent cavities oscillate 180 degrees out of phase with each other or pi radians out of phase. This is known as pi-mode, and is the most power efficient mode. Numerous other modes are possible. For example, all cavities can oscillate in phase with each other, which is known as the zero pi-mode. Another possibility is that adjacent cavities oscillate pi/2 radians or 90 degrees out of phase with each other. In general, the number of distinct possible modes equals the number of resonant cavities. As more cavities are added, the number of possible modes increases.
Without some sort of mode control device, a magnetron can and will oscillate at any possible mode. Each mode has a slightly different oscillation frequency and power efficiency. Without mode control, a magnetron oscillator will jump about in frequency and power level in an uncontrolled manner.
The frequency and power limitations of conventional magnetron designs arise from a breakdown of mode control. Mode control is conventionally accomplished either by using strapping rings 10 as shown in FIGS. 1A and 1B, or by alternating the size of the resonant cavities 12 as in the rising sun design of FIG. 1C. As a practical matter, these prior art methods of mode control fail when the number of cavities exceed approximately twenty. Numbers higher than forty heretofore have been considered completely impractical.
Since the spacing of anode pole pieces depends directly on the operating wavelength, this limitation drives higher frequency designs to very small size and limits their power handling capability. The very small size also requires very large magnetic fields to maintain small radius electron orbits within the small device. At 100 GHz for example, the resonant cavities are reduced to a fraction of a millimeter in length. Such small pieces of metal may cause problems as a result of being unable to handle high-power levels without melting. Furthermore, as the anode diameter becomes smaller, impractically large magnetic fields are required to produce tighter electron orbits around the cathode.
With reference to FIG. 2, a conventional cylindrical magnetron 14 is provided with a central electron emitting cathode 16 and a circumferential anode 18 containing a plurality of resonant cavities 12. A high voltage source (not shown) is used to accelerate electrons from the cathode 16 to the anode 18 (the cathode is at negative potential and the anode is at positive potential), and an axial magnetic field 20 causes the electrons to follow curved orbits on their way from the cathode 16 to the anode 18. A power coupling port 19 provides a means to deliver the energy away from the resonant cavities 12. Planar (non-curved) magnetrons are also possible with similar operating principles. For clarity, only cylindrical magnetrons will be discussed.
During operation of the magnetron 14, an electron cloud rotates about an axis of symmetry within an interaction space, e.g., the space between the anode and cathode. As the cloud rotates, the electron distribution becomes bunched on its outer surface, thereby forming spokes of electronic charge that resemble the teeth on a gear. The operating frequency of the magnetron is determined by how rapidly the spokes pass from one gap to the next in one half of the oscillation period. The electron rotational velocity is determined primarily by the strength of a permanent magnetic field and the electric field which are applied to the interaction region.
FIG. 3 illustrates an expanded view of a portion of a conventional magnetron anode 18 in pi-mode operation. For simplicity, the curved structure is drawn straight. When operating in the desired pi-mode, adjacent resonant cavities 12 oscillate out of phase with each other. The space between the cathode and anode is filled with a rotating electron cloud 22. A high voltage accelerates the electrons from cathode 16 to anode 18 and supplies the electrical energy which is converted into microwave power.
At an instant of time during pi-mode operation, it can be seen that the microwave fringing fields 24 at the resonant cavity openings have alternating directions. The circulating electron cloud 22 sees electric fields across consecutive openings which go from plus to minus potential, then minus to plus, then plus to minus, etc. The result is that the surface of the metal pole pieces 26 between resonant cavity openings are alternately at either positive or negative potential. Since electrons are attracted to positive and repelled from negative potentials, pi-mode operation serves to efficiently bunch the electron cloud 22.
The rotating electron cloud 22 interacts only with the fringing fields 24 between anode poles. The function of the multiplicity of microwave resonators 12 is to support and maintain the oscillating fringing fields 24. As taught in commonly assigned U.S. Pat. No. 6,724,146, a multiplicity of microwave resonators is not necessary to produce magnetron operation. It is sufficient to provide a multiplicity of anode pole pieces that support pi-mode at fringing fields across the anode openings.
For many practical reasons, the distance D between anode openings is typically a fraction of the operating wavelength, such as, for example, one-tenth or one-hundredth of the operating free space wavelength. The anode circumference of a typical prior art microwave-oven magnetron is about one-half the guided-wave wavelength and contains ten resonators for a spacing D of about 1/20 wavelength. It is also known as a practical matter that mode control fails for magnetrons constructed with more than approximately twenty resonant cavities 12. From these two facts it can be seen that mode control is difficult when the circumference of the anode is larger than approximately one wavelength at the operating frequency.
Recently, the applicant has described a high frequency magnetron that is suitable for operating at frequencies heretofore not possible with conventional magnetrons. This high frequency magnetron is capable of producing high efficiency, high power electromagnetic energy at frequencies within the infrared and visible light bands, and which may extend beyond into higher frequency bands such as ultraviolet, x-ray, etc. As a result, the magnetron may serve as a light source in a variety of applications such as long distance optical communications, commercial and industrial lighting, manufacturing, etc. Such magnetron is described in detail in commonly assigned, U.S. Pat. No. 6,373,194 and U.S. Pat. No. 6,504,303, the entire disclosures of which are incorporated herein by reference.
This high frequency magnetron is advantageous as it does not require extremely high magnetic fields. Rather, the magnetron preferably uses a magnetic field of more reasonable strength, and more preferably a magnetic field obtained from permanent magnets. The magnetic field strength determines the radius of rotation and angular velocity of the electron space charge within the interaction region between the cathode and the anode. The anode includes a plurality of small resonant cavities which are sized according to the desired operating wavelength. A mechanism is provided for constraining the plurality of resonant cavities to operate in pi-mode. Specifically, each resonant cavity is constrained to oscillate pi-radians out of phase with the resonant cavities immediately adjacent thereto. An output coupler or coupler array is provided to couple optical radiation away from the resonant cavities in order to deliver useful output power.
Additionally, applicant has made further improvements to the magnetron, wherein the wavelength of operation may be in the microwave band, infrared light or visible light bands, or even shorter wavelengths. The magnetron converts direct current (dc) electricity into single-frequency electromagnetic radiation, and includes an array of phasing lines and/or inter-digitated electrodes that are disposed around the outer circumference of an electron interaction space. During operation, oscillating electric fields appear in gaps between adjacent phasing lines/inter-digitated electrodes in the array. The electric fields are constrained to point in opposite directions in adjacent gaps, thus providing pi-mode fields that are necessary for efficient magnetron operation. Such a magnetron is described in detail in commonly assigned U.S. Pat. No. 6,724,146, the entire disclosure of which is incorporated herein by reference.
Nevertheless, a strong need remains in the art for even further advances in the development of high frequency electromagnetic radiation sources. For example, a strong need remains for a device having improved operation at high frequencies, e.g., over 100 GHz, while operating at high power levels. More particularly, a strong need remains for a device which does not utilize multiple resonant cavities, thereby simplifying the construction of the magnetron. Such a device would offer greater design flexibility and would be particularly well suited for producing electromagnetic radiation at very short wavelengths and operating at high power levels.